G. A. Senthil
G. A. Senthil has a total of 44 co-authors. Here are the top co-authors by number of shared publications:
- R. Prabha
Department of Electronics and Communication Engineering, Sri Sairam Institute of Technology, Chennai, India
14 shared publications - S. Sridevi
Vels Institute of Science, Technology and Advanced Studies
8 shared publications R. Prabha
Sri Sairam Institute of Technology, India
4 shared publications- Monica Komala Manoharan
Vellore Institute of Technology Chennai Campus
3 shared publications - Su. Suganthi
Sri Sai Ram Institute of Technology
3 shared publications - Arun Raaza
Department of Electronics and Communication Engineering, Vels Institute of Science, Technology and Advanced Studies, Chennai, India
2 shared publications
Latest Publications
- 1 September 2025
- Published by Springer Nature
- p. 77-96
A Novel Revolutionizing Approach Sustainable Smart Education Practices for Autism Spectrum Disorder Kids Using Virtual Reality
- 1 September 2025
- Published by Springer Nature
- p. 77-96
A Novel Revolutionizing Approach Sustainable Smart Education Practices for Autism Spectrum Disorder Kids Using Virtual Reality
- Open Access
- 10 April 2025
- Published by EDP Sciences in BIO Web of Conferences
- Vol. 172, 02002
Quantum AI: A Cognitive Machine Learning Technique based on Nurturing Food Security Sustainability Predictive Analysis for Life Science - Bioengineering in Healthcare
Individualized and accurate evaluation of nutrient intake is essential for good health. Disease prevention and increased food security This article combines image analysis with quantum algorithms for precise food insights, introducing an advanced quantum-enhanced AI system. It is designed to predict the nutritional content of foods consumed. The system starts with taking photos of food using a Convolutional Neural Network (CNN) is processed by it has a classification accuracy of 91.87%. User-specific information such as age, weight, height, and BMI are also used to calculate individual nutritional needs. It enables tailored dietary recommendations. Quantum Support Vector Machines (QSVM), Quantum Neural Network (QNN), and Quantum Reinforcement Learning (QRL). The system's Leveraging (QRL) has high prediction accuracies of 90%, 92%, and 93%, ensuring efficient nutritional analysis in different foods. Integrating quantum computer models will greatly improve predictive performance and scalability. This has led to advances in bioengineering applications related to personalized nutrition. The proposed approach has the potential to be widely applied in health care. By helping with personal nutrition planning and supporting nutritional decision-making at a granular level.
- 10 April 2025
- Published by EDP Sciences in BIO Web of Conferences
- Vol. 172, 02002
Open AccessQuantum AI: A Cognitive Machine Learning Technique based on Nurturing Food Security Sustainability Predictive Analysis for Life Science - Bioengineering in Healthcare
Individualized and accurate evaluation of nutrient intake is essential for good health. Disease prevention and increased food security This article combines image analysis with quantum algorithms for precise food insights, introducing an advanced quantum-enhanced AI system. It is designed to predict the nutritional content of foods consumed. The system starts with taking photos of food using a Convolutional Neural Network (CNN) is processed by it has a classification accuracy of 91.87%. User-specific information such as age, weight, height, and BMI are also used to calculate individual nutritional needs. It enables tailored dietary recommendations. Quantum Support Vector Machines (QSVM), Quantum Neural Network (QNN), and Quantum Reinforcement Learning (QRL). The system's Leveraging (QRL) has high prediction accuracies of 90%, 92%, and 93%, ensuring efficient nutritional analysis in different foods. Integrating quantum computer models will greatly improve predictive performance and scalability. This has led to advances in bioengineering applications related to personalized nutrition. The proposed approach has the potential to be widely applied in health care. By helping with personal nutrition planning and supporting nutritional decision-making at a granular level.
- Open Access
- 10 April 2025
- Published by EDP Sciences in BIO Web of Conferences
- Vol. 172, 02001
Integrating Bioengineering and Machine Learning: A Multi-Algorithm Approach to Enhance Agricultural Sustainability and Resource Efficiency
The novel research incorporates high-level machine learning algorithms for optimizing agricultural performance regarding sustainability and resource efficiencies. By using random forests and SVMs, this work successfully achieved 92% prediction accuracy for crop yields and an 89% classification accuracy of agricultural regions, thereby highly enhancing the decision-making power of farmers and policymakers. With over 10,000 historical records, the random forest model established a hypothesis that maize yields could be increased by almost 25% in ideal conditions. At the same time, the SVM identified more strongly within high-productivity areas a yield increase of 15% for targeted crops. Furthermore, Convolutional Neural Networks processed nearly 5,000 satellite images to register a precision rate of up to 94% for early crop stress resulting in a reduction in crop loss by 30%. Reinforcement Learning was used also to reduce water use in irrigation by 20% without impacting the yield of crops while optimizing irrigation schedules to adapt to real-time data concerning the environment toward helping to meet the sustainability goals. Convolutional Neural Network (CNN) stands out as the best algorithm in this context due to its exceptional performance in early detection of crop stress symptoms, achieving 94% accuracy. Findings have indicated that the multi-algorithm approach not only promotes increased predictive capabilities and resource optimization but also raises food safety with the increased threats in agriculture.
- 10 April 2025
- Published by EDP Sciences in BIO Web of Conferences
- Vol. 172, 02001
Open AccessIntegrating Bioengineering and Machine Learning: A Multi-Algorithm Approach to Enhance Agricultural Sustainability and Resource Efficiency
The novel research incorporates high-level machine learning algorithms for optimizing agricultural performance regarding sustainability and resource efficiencies. By using random forests and SVMs, this work successfully achieved 92% prediction accuracy for crop yields and an 89% classification accuracy of agricultural regions, thereby highly enhancing the decision-making power of farmers and policymakers. With over 10,000 historical records, the random forest model established a hypothesis that maize yields could be increased by almost 25% in ideal conditions. At the same time, the SVM identified more strongly within high-productivity areas a yield increase of 15% for targeted crops. Furthermore, Convolutional Neural Networks processed nearly 5,000 satellite images to register a precision rate of up to 94% for early crop stress resulting in a reduction in crop loss by 30%. Reinforcement Learning was used also to reduce water use in irrigation by 20% without impacting the yield of crops while optimizing irrigation schedules to adapt to real-time data concerning the environment toward helping to meet the sustainability goals. Convolutional Neural Network (CNN) stands out as the best algorithm in this context due to its exceptional performance in early detection of crop stress symptoms, achieving 94% accuracy. Findings have indicated that the multi-algorithm approach not only promotes increased predictive capabilities and resource optimization but also raises food safety with the increased threats in agriculture.
- 30 December 2024
- Published by Springer Nature
- p. 118-137
A Novel Analysis and Classification of Pressure Ulcer Prediction for Coma Patients in ICU Using Deep Learning Techniques
- 30 December 2024
- Published by Springer Nature
- p. 118-137
A Novel Analysis and Classification of Pressure Ulcer Prediction for Coma Patients in ICU Using Deep Learning Techniques
- 23 October 2024
- Published by IGI Global Scientific Publishing
- p. 87-102
Quantum Cryptography-Enhanced Cyber Security Intrusion Detection System APTs Attacks in Blockchain
The novel proposed in this paper aims to revolutionize cybersecurity within Blockchain systems by integrating Quantum Cryptography with federated deep reinforcement learning intrusion detection systems (IDPS). This pioneering fusion of cutting-edge technologies offers a multifaceted defense mechanism against advanced persistent threats (APTs) while preserving the decentralized nature of Blockchain networks. Complementing Quantum Cryptography, federated deep reinforcement learning enhances cybersecurity by deploying AI-driven intrusion detection systems across decentralized Blockchain nodes. This decentralized learning paradigm empowers Blockchain networks to adapt dynamically to evolving cyber threats, ensuring timely and effective responses to malicious activities. Quantum Cryptography and federated deep reinforcement learning, the proposed framework defines strategy against sophisticated cyber-attacks, bolstering the resilience of Blockchain systems. Markov Decision Process is the reinforcement learning algorithm used in the proposed system that detects cyber-attacks and threats.
- 23 October 2024
- Published by IGI Global Scientific Publishing
- p. 87-102
Quantum Cryptography-Enhanced Cyber Security Intrusion Detection System APTs Attacks in Blockchain
The novel proposed in this paper aims to revolutionize cybersecurity within Blockchain systems by integrating Quantum Cryptography with federated deep reinforcement learning intrusion detection systems (IDPS). This pioneering fusion of cutting-edge technologies offers a multifaceted defense mechanism against advanced persistent threats (APTs) while preserving the decentralized nature of Blockchain networks. Complementing Quantum Cryptography, federated deep reinforcement learning enhances cybersecurity by deploying AI-driven intrusion detection systems across decentralized Blockchain nodes. This decentralized learning paradigm empowers Blockchain networks to adapt dynamically to evolving cyber threats, ensuring timely and effective responses to malicious activities. Quantum Cryptography and federated deep reinforcement learning, the proposed framework defines strategy against sophisticated cyber-attacks, bolstering the resilience of Blockchain systems. Markov Decision Process is the reinforcement learning algorithm used in the proposed system that detects cyber-attacks and threats.
- 23 October 2024
- Published by Springer Nature
- p. 1-15
Secure Private Data Storage and File Sharing Using Homomorphic and Hybrid Hash-Based Cryptographic Technique in Cloud
- 23 October 2024
- Published by Springer Nature
- p. 1-15
Secure Private Data Storage and File Sharing Using Homomorphic and Hybrid Hash-Based Cryptographic Technique in Cloud
- 1 October 2024
- Published by Springer Nature
- p. 3-20
A Novel Revolutionizing Medical Surgery Procedures Using Mixed Reality
- 1 October 2024
- Published by Springer Nature
- p. 3-20
A Novel Revolutionizing Medical Surgery Procedures Using Mixed Reality